This note is concerned with the formula for diabatic flow in Boer's (1975) paper, where p , is the reference pressure p,(O,t) -%dOdu eT ae u is the area of the earth and 8, is the potential temperature at the top of the atmosphere. For the sake of simplicity, let us aasume that 8 is greater than the potential temperature at any point of the earth's surface so that the integration in (2) can be carried out without difficulty p , (0, t ) is proportional to the mass of the air above the isentropic surface with potential temperature 8. Boer (1975) quotes (1) from Smith (1969) who refers to Leibnitz's rule in order to prove (1).However, it can be shown that (1) is not valid for diabatic flow. Let us suppose that (1) is correct and let us look at a certain time to at the particles a t an isentropic surface with the potential temperature O0. For diabatic flow, 8 = dO/dt + O and the particles will leave the O0surface. Due to ( l ) , all these particles will keep the same reference pressure p,(O0, to). Therefore, all these particles will have the same potential temperature 8(t) for t > t o since p , Tellus XXVIII (1976), 4 (8, t ) is a unique function of 0 and time. Of course, 0 must not coincide with €lo. This is possible only if 8 does not depend on the coordinates z and y. In the general diabatic case, however, the particles will move to different @-surfaces and will have different reference pressures after some time, in contrast to (1).We may express dp,/dt aa a function of 8, p and p,:Integrating the equation of continuity (4) from 8 to 0, and over the area of the earth (8 = 0 for 8 =8,) we obtain -p d a + 8 -d a -0 sa" t 1 : It follows (7)It can be seen from (7) that (1) is not valid, since the first term on the right hand side of ( 7 ) depends on 0 and t only, whereas the second term depends also on z and y (except for 8 =8 (8, t)). Obviously, (1) is correct for adiabatic flow.